CREATIVE INDUSTRIES

Volumetric Video Could Require Up To 2Tbps. Here’s How to Make It Possible

158views

By Caio Castro

What do Star Wars and an elephant in Germany have in common? If you are talking about the holographic elephant at Circus Roncalli, well, it is a lot.

Holograms are becoming more of a reality as time goes by. Circuses like Roncalli in Germany, and L’Écocirque in France, have been replacing live animals with 3D holograms for a while. And as the metaverse develops, this kind of technology tends to become more popular.

Here’s the trick, however: Roncalli’s and L’Écocirque’s 3D animals are combined projections. They are not volumetric and they are not transmitted over the internet – in fact, it will take some time until they are. But you can find people already working to make that a reality.

“In volumetric media, we add the context parallax – multiple views, instead of only two static standpoints,” explained  professor at Ghent University, Filip de Turck during his talk at the IEEE Globecom 2022. “It’s called the six degrees of freedom.”

According to him, huge amount of bandwidth is required to transmit even one volumetric object. In comparison, a standard HD video needs 1.5 Mbps, while a 4K video demands between 15 to 25 Mbps. A 16K video requires 100-500 Mbps.

Now with the volumetric videos: if you want to transmit only one object, it will take 0.5 to 2 Gbps bandwidth. If you include context, like the background, you must provide a staggering 100 Gbps to 2 Tbps connection.

While today’s technology cannot deliver such a type of experience, Turck and a team of researchers are trying to overcome the existing challenges to make volumetric transmissions a reality in the future. They are part of SPIRIT (Scalable Platform for Innovations on Real-time Immersive Telepresence), a consortium funded by European Union’s Horizon Europe launched last October.

“SPIRIT’s goal is to build a telepresence platform,” the professor said. “The three main use cases [we see for it] are holographic conferencing, telesurgery and remote patient monitoring, and remote industrial monitoring and management.”

One of the first steps they took was understanding whether the current transports and applications infrastructure could deal with volumetric media delivery – and to what extent.

The researchers found out that when they favoured the quality of images, areas like processing, buffering, and the protocol overhead would not enable real-time transmission. On the other hand, when they optimised latency, the researchers had a hard time controlling the quality, and the transmission had a very low resilience.

“Users consider the [overall] quality to be too low. That means that current optimisations are insufficient to satisfy the demand,” Turck observed. “These techniques shouldn’t be deployed individually; they must work together. The important question is: how to get the best of both at the same time?”

That’s why the team at SPIRIT is developing a cross-layer approach that considers several factors:

On the end-user side, for instance, optimisations should consider the viewport prediction – that is, a forecast of what users will see in a 360-degree experience. It encompasses visual, audio, and tactile feeds.

A cross-layer approach also synchs those feeds in a continuous, real-time, and accurate way. That’s paramount, according to Turck, to provide a clear Quality of Service and Quality of Experience cybersickness assessment.

Other areas include transport and OTT optimisations, novel buffering techniques, network architectures, and more.

While all of SPIRIT’s work has just started, researchers are already planning to join more events like MWC Barcelona 2023, the IEEE Infocom, and the EUCNC 6G Summit. The next deliverable, its data management plan, is set to be released in January 2023. The project runs through the end of 2025.

Leave a Response

bahis canlı casino siteleri canlı bahis siteleri